Efficient and Concise Explanations for Object Detection with Gaussian-Class Activation Mapping Explainer
This addresses the need for efficient and reliable explanations in object detection for AI practitioners, though it appears incremental as it builds on existing region-based methods.
The paper tackles the challenge of providing quick and plausible explanations for object detection models in Explainable AI by introducing the Gaussian Class Activation Mapping Explainer (G-CAME), which reduces explanation time to 0.5 seconds while maintaining quality and reducing bias in tiny object detection.
To address the challenges of providing quick and plausible explanations in Explainable AI (XAI) for object detection models, we introduce the Gaussian Class Activation Mapping Explainer (G-CAME). Our method efficiently generates concise saliency maps by utilizing activation maps from selected layers and applying a Gaussian kernel to emphasize critical image regions for the predicted object. Compared with other Region-based approaches, G-CAME significantly reduces explanation time to 0.5 seconds without compromising the quality. Our evaluation of G-CAME, using Faster-RCNN and YOLOX on the MS-COCO 2017 dataset, demonstrates its ability to offer highly plausible and faithful explanations, especially in reducing the bias on tiny object detection.